An L-2-Boosting Algorithm for Estimation of a Regression Function
- Bagirov, Adil, Clausen, Conny, Kohler, Michael
- Authors: Bagirov, Adil , Clausen, Conny , Kohler, Michael
- Date: 2010
- Type: Text , Journal article
- Relation: IEEE Transactions on Information Theory Vol. 56, no. 3 (2010), p. 1417-1429
- Full Text:
- Reviewed:
- Description: An L-2-boosting algorithm for estimation of a regression function from random design is presented, which consists of fitting repeatedly a function from a fixed nonlinear function space to the residuals of the data by least squares and by defining the estimate as a linear combination of the resulting least squares estimates. Splitting of the sample is used to decide after how many iterations of smoothing of the residuals the algorithm terminates. The rate of convergence of the algorithm is analyzed in case of an unbounded response variable. The method is used to fit a sum of maxima of minima of linear functions to a given data set, and is compared with other nonparametric regression estimates using simulated data.
- Authors: Bagirov, Adil , Clausen, Conny , Kohler, Michael
- Date: 2010
- Type: Text , Journal article
- Relation: IEEE Transactions on Information Theory Vol. 56, no. 3 (2010), p. 1417-1429
- Full Text:
- Reviewed:
- Description: An L-2-boosting algorithm for estimation of a regression function from random design is presented, which consists of fitting repeatedly a function from a fixed nonlinear function space to the residuals of the data by least squares and by defining the estimate as a linear combination of the resulting least squares estimates. Splitting of the sample is used to decide after how many iterations of smoothing of the residuals the algorithm terminates. The rate of convergence of the algorithm is analyzed in case of an unbounded response variable. The method is used to fit a sum of maxima of minima of linear functions to a given data set, and is compared with other nonparametric regression estimates using simulated data.
Exercise, mood, self-efficacy, and social support as predictors of depressive symptoms in older adults : Direct and interaction effects
- Miller, Kyle, Mesagno, Christopher, McLaren, Suzanne, Grace, Fergal, Yates, Mark, Gomez, Rapson
- Authors: Miller, Kyle , Mesagno, Christopher , McLaren, Suzanne , Grace, Fergal , Yates, Mark , Gomez, Rapson
- Date: 2019
- Type: Text , Journal article
- Relation: Frontiers in Psychology Vol. 10, no. (2019), p. 1-11
- Full Text:
- Reviewed:
- Description: Background: Depression is a chronic condition that affects up to 15% of older adults. The healthogenic effects of regular exercise are well established, but it is still unclear which exercise-related variables characterise the antidepressant effects of exercise. Thus, the purpose of this study was to examine the extent to which exercise-related variables (exercise behaviour, exercise-induced mood, exercise self-efficacy, and social support) can predict depressive symptoms in a cohort of community-dwelling older adults. Methods: This study employed a cross-sectional analysis of questionnaire data from a sample of 586 community-dwelling older Australians aged 65 to 96 years old. Participants completed the Center for Epidemiologic Studies Depression Scale, modified CHAMPS Physical Activity Questionnaire for Older Adults, Four-Dimension Mood Scale, Self-Efficacy for Exercise Scale, and Social Provisions Scale - Short Form. Bivariate correlations were performed, and hierarchical multiple regression was subsequently used to test the regression model. Results: Exercise behaviour, exercise-induced mood, exercise self-efficacy, and social support were all negatively associated with depressive symptoms (r = -0.20 to -0.56). When the variables were entered as predictors into the hierarchical multiple regression model, social support was the strongest predictor of depressive symptoms (beta = -0.42), followed by exercise-induced mood (beta = -0.23), and exercise self-efficacy (beta = -0.07). Exercise behaviour did not explain any additional variance in depressive symptoms. A modest interaction effect was also observed between exercise-induced mood and social support. Conclusion: These findings indicate that social support is the strongest predictor of depressive symptomology in community-dwelling older adults, particularly when combined with positive exercise-induced mood states. When addressing the needs of older adults at risk of depression, healthcare professionals should consider the implementation of exercise programmes that are likely to benefit older adults by improving mood, enhancing self-efficacy, and building social support.
- Authors: Miller, Kyle , Mesagno, Christopher , McLaren, Suzanne , Grace, Fergal , Yates, Mark , Gomez, Rapson
- Date: 2019
- Type: Text , Journal article
- Relation: Frontiers in Psychology Vol. 10, no. (2019), p. 1-11
- Full Text:
- Reviewed:
- Description: Background: Depression is a chronic condition that affects up to 15% of older adults. The healthogenic effects of regular exercise are well established, but it is still unclear which exercise-related variables characterise the antidepressant effects of exercise. Thus, the purpose of this study was to examine the extent to which exercise-related variables (exercise behaviour, exercise-induced mood, exercise self-efficacy, and social support) can predict depressive symptoms in a cohort of community-dwelling older adults. Methods: This study employed a cross-sectional analysis of questionnaire data from a sample of 586 community-dwelling older Australians aged 65 to 96 years old. Participants completed the Center for Epidemiologic Studies Depression Scale, modified CHAMPS Physical Activity Questionnaire for Older Adults, Four-Dimension Mood Scale, Self-Efficacy for Exercise Scale, and Social Provisions Scale - Short Form. Bivariate correlations were performed, and hierarchical multiple regression was subsequently used to test the regression model. Results: Exercise behaviour, exercise-induced mood, exercise self-efficacy, and social support were all negatively associated with depressive symptoms (r = -0.20 to -0.56). When the variables were entered as predictors into the hierarchical multiple regression model, social support was the strongest predictor of depressive symptoms (beta = -0.42), followed by exercise-induced mood (beta = -0.23), and exercise self-efficacy (beta = -0.07). Exercise behaviour did not explain any additional variance in depressive symptoms. A modest interaction effect was also observed between exercise-induced mood and social support. Conclusion: These findings indicate that social support is the strongest predictor of depressive symptomology in community-dwelling older adults, particularly when combined with positive exercise-induced mood states. When addressing the needs of older adults at risk of depression, healthcare professionals should consider the implementation of exercise programmes that are likely to benefit older adults by improving mood, enhancing self-efficacy, and building social support.
Non-functional regression : A new challenge for neural networks
- Vamplew, Peter, Dazeley, Richard, Foale, Cameron, Choudhury, Tanveer
- Authors: Vamplew, Peter , Dazeley, Richard , Foale, Cameron , Choudhury, Tanveer
- Date: 2018
- Type: Text , Journal article
- Relation: Neurocomputing Vol. 314, no. (2018), p. 326-335
- Full Text:
- Reviewed:
- Description: This work identifies an important, previously unaddressed issue for regression based on neural networks – learning to accurately approximate problems where the output is not a function of the input (i.e. where the number of outputs required varies across input space). Such non-functional regression problems arise in a number of applications, and can not be adequately handled by existing neural network algorithms. To demonstrate the benefits possible from directly addressing non-functional regression, this paper proposes the first neural algorithm to do so – an extension of the Resource Allocating Network (RAN) which adds additional output neurons to the network structure during training. This new algorithm, called the Resource Allocating Network with Varying Output Cardinality (RANVOC), is demonstrated to be capable of learning to perform non-functional regression, on both artificially constructed data and also on the real-world task of specifying parameter settings for a plasma-spray process. Importantly RANVOC is shown to outperform not just the original RAN algorithm, but also the best possible error rates achievable by any functional form of regression.
- Authors: Vamplew, Peter , Dazeley, Richard , Foale, Cameron , Choudhury, Tanveer
- Date: 2018
- Type: Text , Journal article
- Relation: Neurocomputing Vol. 314, no. (2018), p. 326-335
- Full Text:
- Reviewed:
- Description: This work identifies an important, previously unaddressed issue for regression based on neural networks – learning to accurately approximate problems where the output is not a function of the input (i.e. where the number of outputs required varies across input space). Such non-functional regression problems arise in a number of applications, and can not be adequately handled by existing neural network algorithms. To demonstrate the benefits possible from directly addressing non-functional regression, this paper proposes the first neural algorithm to do so – an extension of the Resource Allocating Network (RAN) which adds additional output neurons to the network structure during training. This new algorithm, called the Resource Allocating Network with Varying Output Cardinality (RANVOC), is demonstrated to be capable of learning to perform non-functional regression, on both artificially constructed data and also on the real-world task of specifying parameter settings for a plasma-spray process. Importantly RANVOC is shown to outperform not just the original RAN algorithm, but also the best possible error rates achievable by any functional form of regression.
Livelihood activities and skills in rural areas of the Zambezi Region, Namibia : Implications for policy and poverty reduction
- Kamwi, Jonathan, Chirwa, Paxie, Graz, Patrick, Manda, Samuel, Mosimane, Alfons, Katsch, Christoph
- Authors: Kamwi, Jonathan , Chirwa, Paxie , Graz, Patrick , Manda, Samuel , Mosimane, Alfons , Katsch, Christoph
- Date: 2018
- Type: Text , Journal article
- Relation: African Journal of Food, Agriculture, Nutrition and Development Vol. 18, no. 1 (2018), p. 13074-13094
- Full Text:
- Reviewed:
- Description: This paper examined livelihood activities and skill sets available within rural households in the Zambezi Region of Namibia. Specifically, the study addressed three key questions: (i) what livelihood activities do rural people pursue? (ii) what demographic factors are associated with these activities? and (iii) what measures can be taken to diversify and sustain income from these livelihood activities? In order to address these questions, semistructured interviews covering 424 households were used to collect the data. The questionnaire consisted of questions corresponding to the sustainable livelihood framework including (1) human assets (2) financial assets and major sources of income (3) physical and natural assets and (4) social assets. A series of logistic regressions were fitted from which the estimated odds ratios (y) were derived to ascertain the effect of the predictors on the livelihood activities and skills. Odds ratios were used to measure the magnitude of strength of association or non-independence between binary data values. The results showed that the use of various livelihood activities and skills in different combinations is of significant importance to rural livelihoods. Five percent of the respondents obtained income from only one source, with 95 % of the respondents engaged in a combination of farming and non-farming activities. Most of the respondents had various reasons for diversifying into other activities vis-a-vis agricultural income, limited skills, large family size, availability of opportunities, seasonal nature of agricultural produce, favourable demand for goods and services or a combination of these. In addition, the results showed that gender, age, designation and education significantly (p < 0.05) influenced the choice of household's skills. The study concludes that a combination of rural household activities and skills influenced by a variety of factors have led to improved livelihoods in the study area. For policy purposes, this suggests that state interventions in rural livelihood skill development can play a significant role in promoting more sustainable rural livelihoods. © 2018, African Scholarly Science Communications Trust (ASSCAT).
- Authors: Kamwi, Jonathan , Chirwa, Paxie , Graz, Patrick , Manda, Samuel , Mosimane, Alfons , Katsch, Christoph
- Date: 2018
- Type: Text , Journal article
- Relation: African Journal of Food, Agriculture, Nutrition and Development Vol. 18, no. 1 (2018), p. 13074-13094
- Full Text:
- Reviewed:
- Description: This paper examined livelihood activities and skill sets available within rural households in the Zambezi Region of Namibia. Specifically, the study addressed three key questions: (i) what livelihood activities do rural people pursue? (ii) what demographic factors are associated with these activities? and (iii) what measures can be taken to diversify and sustain income from these livelihood activities? In order to address these questions, semistructured interviews covering 424 households were used to collect the data. The questionnaire consisted of questions corresponding to the sustainable livelihood framework including (1) human assets (2) financial assets and major sources of income (3) physical and natural assets and (4) social assets. A series of logistic regressions were fitted from which the estimated odds ratios (y) were derived to ascertain the effect of the predictors on the livelihood activities and skills. Odds ratios were used to measure the magnitude of strength of association or non-independence between binary data values. The results showed that the use of various livelihood activities and skills in different combinations is of significant importance to rural livelihoods. Five percent of the respondents obtained income from only one source, with 95 % of the respondents engaged in a combination of farming and non-farming activities. Most of the respondents had various reasons for diversifying into other activities vis-a-vis agricultural income, limited skills, large family size, availability of opportunities, seasonal nature of agricultural produce, favourable demand for goods and services or a combination of these. In addition, the results showed that gender, age, designation and education significantly (p < 0.05) influenced the choice of household's skills. The study concludes that a combination of rural household activities and skills influenced by a variety of factors have led to improved livelihoods in the study area. For policy purposes, this suggests that state interventions in rural livelihood skill development can play a significant role in promoting more sustainable rural livelihoods. © 2018, African Scholarly Science Communications Trust (ASSCAT).
A novel OFDM format and a machine learning based dimming control for lifi
- Nowrin, Itisha, Mondal, M., Islam, Rashed, Kamruzzaman, Joarder
- Authors: Nowrin, Itisha , Mondal, M. , Islam, Rashed , Kamruzzaman, Joarder
- Date: 2021
- Type: Text , Journal article
- Relation: Electronics (Switzerland) Vol. 10, no. 17 (2021), p.
- Full Text:
- Reviewed:
- Description: This paper proposes a new hybrid orthogonal frequency division multiplexing (OFDM) form termed as DC‐biased pulse amplitude modulated optical OFDM (DPO‐OFDM) by combining the ideas of the existing DC‐biased optical OFDM (DCO‐OFDM) and pulse amplitude modulated discrete multitone (PAM‐DMT). The analysis indicates that the required DC‐bias for DPO‐OFDM-based light fidelity (LiFi) depends on the dimming level and the components of the DPO‐OFDM. The bit error rate (BER) performance and dimming flexibility of the DPO‐OFDM and existing OFDM schemes are evaluated using MATLAB tools. The results show that the proposed DPO‐OFDM is power efficient and has a wide dimming range. Furthermore, a switching algorithm is introduced for LiFi, where the individual components of the hybrid OFDM are switched according to a target dimming level. Next, machine learning algorithms are used for the first time to find the appropriate proportions of the hybrid OFDM components. It is shown that polynomial regression of degree 4 can reliably predict the constellation size of the DCO‐OFDM component of DPO‐OFDM for a given constellation size of PAM‐DMT. With the component switching and the machine learning algorithms, DPO‐OFDM‐based LiFi is power efficient at a wide dimming range. © 2021 by the authors. Licensee MDPI, Basel, Switzerland.
- Authors: Nowrin, Itisha , Mondal, M. , Islam, Rashed , Kamruzzaman, Joarder
- Date: 2021
- Type: Text , Journal article
- Relation: Electronics (Switzerland) Vol. 10, no. 17 (2021), p.
- Full Text:
- Reviewed:
- Description: This paper proposes a new hybrid orthogonal frequency division multiplexing (OFDM) form termed as DC‐biased pulse amplitude modulated optical OFDM (DPO‐OFDM) by combining the ideas of the existing DC‐biased optical OFDM (DCO‐OFDM) and pulse amplitude modulated discrete multitone (PAM‐DMT). The analysis indicates that the required DC‐bias for DPO‐OFDM-based light fidelity (LiFi) depends on the dimming level and the components of the DPO‐OFDM. The bit error rate (BER) performance and dimming flexibility of the DPO‐OFDM and existing OFDM schemes are evaluated using MATLAB tools. The results show that the proposed DPO‐OFDM is power efficient and has a wide dimming range. Furthermore, a switching algorithm is introduced for LiFi, where the individual components of the hybrid OFDM are switched according to a target dimming level. Next, machine learning algorithms are used for the first time to find the appropriate proportions of the hybrid OFDM components. It is shown that polynomial regression of degree 4 can reliably predict the constellation size of the DCO‐OFDM component of DPO‐OFDM for a given constellation size of PAM‐DMT. With the component switching and the machine learning algorithms, DPO‐OFDM‐based LiFi is power efficient at a wide dimming range. © 2021 by the authors. Licensee MDPI, Basel, Switzerland.
Regression modelling for prediction of clogging in non-vegetated stormwater filters
- Meade, Ben, Khorshidi, Hadi, Kandra, Harpreet, Barton, Andrew
- Authors: Meade, Ben , Khorshidi, Hadi , Kandra, Harpreet , Barton, Andrew
- Date: 2018
- Type: Text , Conference paper
- Relation: 10th International Conference on Water Sensitive Urban Design: Creating water sensitive communities (WSUD 2018 & Hydropolis 2018), 12-15 February 2018, Perth, Western Australia p. 8
- Full Text:
- Reviewed:
- Authors: Meade, Ben , Khorshidi, Hadi , Kandra, Harpreet , Barton, Andrew
- Date: 2018
- Type: Text , Conference paper
- Relation: 10th International Conference on Water Sensitive Urban Design: Creating water sensitive communities (WSUD 2018 & Hydropolis 2018), 12-15 February 2018, Perth, Western Australia p. 8
- Full Text:
- Reviewed:
- «
- ‹
- 1
- ›
- »